Anda di halaman 1dari 26

CHANGES.txt - 08/01/2006 -----------------------CHANGES IN HTMLDOC v1.8.

27 BUG FIXES - Fixed a crash bug that appeared when more than 10 blank pages were present in a document (PR #6223) - Color changes were not reflected in PRE text (STR #129) - Remote URLs did not always work on older operating systems (PR #6179) - Image filenames using % escapes were not decoded properly. - Rows using BGCOLOR that spanned across multiple pages did not render properly (PR #6149) - Rows no longer start on a new page due to a cell with both HEIGHT and ROWSPAN specified (PR #6086) - CMYK JPEG images caused HTMLDOC to crash (PR #6106) - Table cell width calculations didn't always account for the proper minimum width (STR #121) - Images were not copied when generating indexed HTML output to a directory (STR #117) - Changing the bottom margin resulted in text that was formatted below the bottom margin. - The Monospace-Oblique font was not embedded properly in PDF files. CHANGES IN HTMLDOC v1.8.26 BUG FIXES - Outline and keyword strings in PDF files are now stored as Unicode (STR #113) - The Flate compression code could get in an infinite loop if it ran out of memory (STR #101) - Book files saved from the GUI did not handle filenames with spaces (STR #105) - Fixed and re-enabled the ASCII85Device filter support in PostScript Level 2/3 output (STR #116) - Character entities in the first word of a file were not rendered properly (STR #111) - Fixed-size table columns were incorrectly resized when a table width was also specified and there was extra space to distribute (PR #6062) - Text could "walk" up or down when in-line images were used (PR #6034) - Row backgrounds incorrectly replaced cell backgrounds when the first cell in a row used ROWSPAN (PR #6033, PR #6036) - HTMLDOC did not correctly parse FONT FACE attributes (PR #6006) - Images in Level 2/3 PostScript output did not work on some printers (PR #6000) - The GUI did not use the first page header (PR #5978) CHANGES IN HTMLDOC v1.8.25

NEW FEATURES - Added "--overflow" and "--no-overflow" command-line options to show or hide the content-too-large errors; the default is "--no-overflow". - Added "--header1" command-line option and "HEADER1" page comments to set the page header for the first page of each chapter. - Added "timing" and "remotebytes" debug data generation. - Added DejaVu font collection to better support Cyrillic and Greek text; the new fonts are available under the generic names "monospace", "sans", and "serif". - Added "--referer" command-line option and corresponding CGI-mode support to pass Referer: information in HTTP requests (STR #91) CHANGES: - On Windows, HTMLDOC now logs CGI mode errors to a file called "htmldoc.log" in the Windows temporary directory. - HTMLDOC no longer uses Base-85 encoding for image data when producing Level 2 and 3 PostScript output. It appears that many printers and PostScript interpreters cannot properly decode this data when the original image data is not a multiple of 8 bits. - HTMLDOC now renders STRONG elements in boldface instead of bold-italic to match the W3C recommendations. - HTMLDOC now automatically inserts a TR element before a TD or TH element as needed to improve web site compatibility; this also triggers a HTML error in --strict mode. BUG FIXES "$HFIMAGEn" didn't work in a header/footer string. HTMLDOC could crash when rendering a table. Book files were not used in CGI mode (STR #69) Cookies were not sent in HTTP requests (STR #71) Table cells were not aligned properly when the ROWSPAN attribute was set to 1 (STR #73) HTMLDOC crashed when rendering unresolved hyperlinks in aligned images (STR #62) Documented the HTMLDOC_NOCGI environment variable (STR #63) HTMLDOC sometimes crashed when rendering tables with background colors (STR #65) HTMLDOC would crash when writing encrypted strings longer than 1024 bytes (STR #66) HTMLDOC didn't set the data directory when running in CGI mode on Windows. HTMLDOC could crash when loading the Symbol.afm file (STR #93) HTMLDOC did not always honor HEIGHT attributes in table rows. Tables with a mix of colspan and rowspan sometimes caused cells to be moved vertically outside the cell.


NEW FEATURES - HTMLDOC now provides limited cookie support via the "--cookies" command-line option and via the cookies passed by a browser in CGI mode. - HTMLDOC now features a CGI mode which provides PDF conversion functionality for web servers. - HTMLDOC now generates a document outline for each input file or URL in webpage mode; the outline shows the title for the file and links to the first page containing that file. - HTMLDOC now offers an "htmlsep" output type which generates HTML output with a separate file for each heading in the table of contents. - HTMLDOC now includes LINK elements in generated HTML so that intelligent browsers like Mozilla can show next/prev/contents/top buttons. - HTMLDOC now supports the BORDERCOLOR attribute for tables, a MSIE extension. - The "strict HTML" mode now reports unresolved local links. - Added support for HP LaserJet 5000 and Xerox DocuPrint 2000/100 printer commands. - Added multiple header/footer image support. - Links to external URLs are now resolved so that the output file can be moved without affecting them. CHANGES - The command-line now allows --fontsize values from 4 to 26 to match the GUI. - Now use a 0.001 point tolerance when checking for content that overflows the page/cell. - HTMLDOC no longer enables interpolation of 2-color images. - The default vertical alignment of images is "BOTTOM" to match the HTML specification. - Paragraph spacing is only applied to the first table after a paragraph. - The tabloid media size was 10 points too short in length. - The table formatter now subtracts the outside border and padding widths for percentage-based widths. This helps to eliminate "truncation or overlapping" errors. - Dropped support for FLTK 1.0.x when building the GUI. - The default vertical alignment is now "bottom" inside paragraphs to correctly align different sized text and images to the baseline. - Indexed images are now written as PDF image objects when encryption is enabled; this works around a serious bug in Acrobat 6 which tries to decrypt the colormap of in-line images twice, causing some very strange colors! - Table captions can now be bottom aligned. - Blocks now break at the bottom of a page if the current line height + standard line height goes below the bottom of the page; this prevents images with captions from getting erroneously moved to the top of the next page.

- Character entities are now supported in HTML attributes and unknown or invalid character entities are left as plain text. - Changed handling of NOWRAP for some tables. - The --permissions option now supports multiple permission keywords in a single invocation. - Dropped support for MacOS 9 and earlier. - HTMLDOC now breaks between images that are too large to fit on a single line, to match the behavior of Mozilla/Netscape (STR #7). - HTMLDOC now handles XHTML input more cleanly. - HTMLDOC no longer specifies an interpolation preference for images in PostScript or PDF output (STR #8) - The DT element no longer applies an italic style (PR #5178) - HTMLDOC now ignores content inside a STYLE element (PR #5183) BUG FIXES - Switching between landscape and portrait orientations would cause margin creepage. - Images did not default to align=bottom, and the align=bottom line spacing calculation was incorrect. - Whitespace before a link was underlined. - Fixed a table column sizing bug. - HTMLDOC didn't read back the HTTP response properly in all situations. - Fixed some more PNG transparency cases. - The PageBoundingBox comments in PostScript output did not account for the back page when duplexing was enabled. - HTMLDOC generated an incorrect image mask for some images. - The first page of each chapter did not use the custom page number if it was placed inside the heading. - HTMLDOC did not reset the rendering cache before each page when producing N-up output; this caused font errors in some cases that prevented the document from printing or displaying properly. - Eliminated a common cause of "table too wide" formatting errors, - Fixed a bug when applying a table background color to a cell without a border that cross a page boundary. - Fixed some calls to strcpy with overlapping arguments. - The names object was never set when the name objects were written. - Character entities were not decoded/encoded inside HTML comments. - The current heading was not always correctly substituted when used in the page header or footer. - When converting web pages from the GUI, the table-of-contents page number preferences were incorrectly used. - PDF page effects/transitions were not put in the right part of the page dictionary, causing them not to be used by the PDF reader application. - The _HD_OMIT_TOC attribute was not being honored for HTML output.

- HTMLDOC now handles "open" messages from the MacOS X Finder (STR #3) - The GUI did not load or save the "strict HTML" setting (STR #6) - The HTML version of the title page did not set the ALT attribute for the title image (STR #10) - The HTML version of the table of contents did not correctly nest the lists in the parent items (STR #10) - Borders around left and right-aligned images were not drawn properly (PR #5112) - Grayscale PDF output was not truly grayscale (STR #32) - Fixed a table-of-contents bug introduced in 1.8.24rc1 which caused the PDF document outline and actual TOC pages were not rendered properly (STR #37) - Links were not rendered due to a bug that was introduced in 1.8.24rc2 (STR #41) CHANGES IN HTMLDOC v1.8.23 NEW FEATURES - HTMLDOC now supports a full alpha channel in PNG files. - HTMLDOC now reports an error when a table, image, or section of text overflows into an adjacent table cell or off the right edge of the page. CHANGES - The NEW SHEET page comment now breaks on N-up boundaries when N is greater than 1. BUG FIXES - HTMLDOC tried to format tables with no rows or columns. While the HTML in technically not in error, it is not exactly something you'd expect someone to do. - HTMLDOC didn't report an error when it could not find the specified title page file. - HTMLDOC could crash if it was unable to create its output files. - HTMLDOC could crash when writing HTML output containing unknown HTML elements. - HTMLDOC could crash when writing HTML output if the output document had no title. - The htmlGetText() function used a fixed-size (10k) buffer which allowed for a buffer overflow. The new code (from HTMLDOC 1.9) allocates its buffer instead. - The header/footer text was not centered properly if the header/footer font size was different than the default body font size. - The GUI interface incorrectly localized URLs when doing a "save as" operation. - The PNG background color was not correct for PNG files using <= 8 bits per pixel. - The HTML parser didn't close the current list item if an intermediate block element (P, PRE, etc.) was inside the previous, unterminated list item.

CHANGES IN HTMLDOC v1.8.22 NEW FEATURES - Now support many Windows code pages in addition to ISO charsets. BUG FIXES - HTMLDOC could crash when checking if a URL is already cached. - HTMLDOC didn't adjust the top margin when changing the page header if the comment didn't appear at the top of a page. - HTMLDOC didn't initialize the right number of TOC headings. - When using a logo image in the header, the header was placed too low on the page. CHANGES IN HTMLDOC v1.8.21 NEW FEATURES - HTMLDOC now supports heading levels 1 to 15. - HTMLDOC now allows the author to omit headings from the TOC using the _HD_OMIT_TOC attribute. - HTMLDOC now supports remote book files when running from the command-line. - HTMLDOC now supports hexadecimal character constants (&#xFF) CHANGES - HTMLDOC now calculates the resolution of the body image using the printable width instead of the page width. - HTMLDOC should now compile out-of-the-box using the Cygwin tools. - HTMLDOC no longer inserts whitespace between text inside DIV elements. - HTMLDOC now supports quoted usernames and passwords in URLs. - HTMLDOC now defaults unknown colors to white for background colors and black for foreground colors. This should make documents that use non-standard color names still appear readable. BUG FIXES - "make install" didn't work in the fonts directory. - "&euro;" didn't work, while "&#128;" did: the character name table was not sorted properly... - Links didn't always point to the right page in PDF output. - XRX comment output could crash HTMLDOC. - Fixed-width columns in tables could be resized by HTMLDOC.

- When writing PostScript commands, some printers reset their duplexing state when a new setpagedevice command is received; we now cache the current duplex state and change it only as needed. - The MEDIA SIZE comment didn't adjust the printable size for the current landscape setting. - HTMLDOC placed the header one line too high. - When continuing a chapter onto the next page, H3 and higher headings would be indented the wrong amount. CHANGES IN HTMLDOC v1.8.20 NEW FEATURES - New --nup and NUMBER-UP options for PostScript and PDF output. - HTMLDOC now logs HTML errors. - HTMLDOC now supports the A3, B, Legal, and Tabloid size names. - HTMLDOC now supports embedding of the base Type1 fonts in PostScript and PDF output. CHANGES - The HTML parser now allows BODY to auto-close HEAD and visa-versa. BUG FIXES - HTMLDOC wouldn't compile using GCC under HP-UX due to a badly "fixed" system header file (vmtypes.h). - Generating a book without a table-of-contents would produce a bad PDF file. - The Xerox XRX comments used the wrong units for the media size, points instead of millimeters. - IMG elements with links that use the ALIGN attribute didn't get the links. - Header and footer comments would interfere with the top and bottom margin settings. - Fixed a bug in the htmlReadFile() function which caused user-provided title pages not to be displayed in PS or PDF output. - The table-of-contents would inherit the last media settings in the document, but use the initial settings when formatting. CHANGES IN HTMLDOC v1.8.19 NEW FEATURES - Now support the "subject" meta variable. CHANGES - Updated the HTML parser to use HTML 4.0 rules for embedding elements inside a LI. - Now check for a TYPE attribute on EMBED elements, so that embedded Flash files do not get treated as HTML.

- Now put the COPYRIGHT meta data in the Author field in a PDF file along with the AUTHOR meta data (if any). - No longer embed the command header when PostScript commands are not being embedded in the output. - HTMLDOC now properly ignores the HTML 4.0 COL element. BUG FIXES - Squeezed tables were not centered or right-aligned properly. - Cells didn't align properly if they were the first things on the page, or if there were several intervening empty cells. - The preferred cell width handling didn't account for the minimum cell width, which could cause some tables to become too large. - Remote URLs didn't always resolve properly (like the images from the Google web page...) - The font width loading code didn't force the non-breaking space to have the same width as a regular space. - PRE text didn't adjust the line height for the tallest fragment in the line. - HTMLDOC tried to seek backwards when reading HTML from the standard input. - The media margin comments did not work properly when the current media orientation was landscape. CHANGES IN HTMLDOC v1.8.18 NEW FEATURES - Added support for remote HTML title pages. CHANGES - Now accept all JPEG files, even if they don't start with an APPn marker. - Now only start a new page for a chapter/filter if we aren't already at the top of a page. BUG FIXES - ROWSPAN handling in tables has been updated to match the MSIE behavior, where the current rowspan is reduced by the minimum rowspan in the table; that is, if you use "ROWSPAN=17" for all cells in a row, HTMLDOC now treats this as if you did not use ROWSPAN at all. It is unclear if this is what the W3C intends. - The "--webpage" option didn't force toc levels to 0, which caused a bad page object reference to be inserted in the PDF output file. - Background colors in nested tables didn't always get drawn in the right order, resulting in the wrong colors showing through. - The HEADER page comment didn't set the correct top position in landscape orientation.

CHANGES IN HTMLDOC v1.8.17 NEW FEATURES - Improved table-of-contents generation, with chapter headings at the top of new TOC pages and page numbers based on the header/footer string. - Added new "--no-localfiles" option to disable access to local files for added security in web services. - Long lines in book files can not be broken up using a trailing backslash. - Added a modern "skin" to the GUI interface. CHANGES - Made some changes in how COLSPAN and ROWSPAN are handled to better match how Netscape and MSIE format things. - HTMLDOC now handles .book files with CR, LF, or CR LF line endings. - Changed the TOC numbering to use 32-bit integers instead of 8-bit integers... - Now handle local links with quoted (%HH) characters. - The command-line interface no longer sets PDF output mode when using --continuous or --webpage. - HTMLDOC now opens HTML output files in binary mode to prevent extra CR's under Windows, and strips incoming CR's from PRE text. - Now support inserting the current chapter and heading in the table-of-contents headers and footers. BUG FIXES - The table cell border and background were offset by the cellpadding when they should only be offset by the cellspacing. - The buffer used for periods that lead up to the page number in the table-of-contents was not large enough for a legal-size document in landscape format. - If a book only contained chapter headings, the PDF bookmarks would be missing the last chapter heading. - Table cells that ended with a break would render incorrectly. - Fixed the table pre-format sizing code to properly account for borders, padding, etc. - Fixed the table squeezing code to honor minimum widths and properly resize the remaining space. - The MEDIA SIZE page comment did not reset the printable width and length of the page. - Tables that used COLSPAN did not honor WIDTH values in non-spanned cells. CHANGES IN HTMLDOC v1.8.16 CHANGES - Now break before and after DIV groups to match most

browsers (the HTML spec is ambivalent about it...) BUG FIXES - HR elements didn't render properly. - Background images didn't render properly and could lock up HTMLDOC. - The "HALF PAGE" comment would lock up HTMLDOC HTMLDOC would keep adding pages until it ran out of memory. - SUP and SUB used a fixed (reduced) size instead of using a smaller size from the current one. - Empty cells could cause unnecessary vertical alignment on the same row. CHANGES IN HTMLDOC v1.8.15 NEW FEATURES - Now support media source, type, and color attributes in PS output. - Now support per-page size, margins, headers, footers, orientation, and duplexing. - Now support plain text for headers and footers, with $ variables to include page numbers and so forth. - New device control prolog file for printer-specific option commands. - Now support a new continuous web page mode that doesn't automatically insert a page break with each HTML file or URL (--continuous). - Now draw border around inline images as needed. - Now support MacOS X (only command-line at present). - Now support the "page-break-before", "text-align", "vertical-align" style attributes, but only for style information in an element's STYLE attribute. CHANGES - Now load images into memory only as needed, and unload them when no longer needed. This provides a dramatic reduction in memory usage with files that contain a lot of in-line images. - Now use the long names for the Flate and DCT filters in all non-inline PDF streams. This avoids a stupid bug in Acrobat Reader when printing to PostScript printers. - HTMLDOC now strips any trailing GET query information when saving the start of files (target) in a document. - Unqualified URLs (no leading scheme name, e.g. http:) now default to the HTTP port (80) instead of the IPP port (631). - Optimized the image writing code to do more efficient color searching. This provides a significant speed improvement when including images. - Now hide all text inside SCRIPT, SELECT, and TEXTAREA elements. - OS/2 port changes from Alexander Mai. BUG FIXES

- If a document started with a heading greater than H1, HTMLDOC would crash. - Full justification would incorrectly be applied to text ending with a break. - Images using ALIGN="MIDDLE" were not centered properly on the baseline. - Table cells that used both ROWSPAN and COLSPAN did not format properly (the colspan was lost after the first row.) - Tables that used cells that exclusively used COLSPAN did not format properly. - When writing HTML output, image references would incorrectly be mapped using the current path. - Images with a width or height of 0 should not be written to PS or PDF output. - The CreationDate comment in PostScript output contained a bad timezone offset (+-0500, for example, instead of -0500). - The PHP portal example now verifies that the URL passed to it contains no illegal characters. CHANGES IN HTMLDOC v1.8.14 NEW FEATURES - Added support for 128-bit encryption. - Added support for GET form request data in the PHP and Java "portal" examples. CHANGES - Most output generation limits have been removed; HTMLDOC now dynamically allocates memory as needed for pages, images, headings, and links. This has the happy side-effect of reducing the initial memory footprint significantly. - Now call setlocale() when it is available to localize the date and time in the output. - The table parsing code now checks to see that a ROWSPAN attribute fits in the table; e.g., a ROWSPAN of 10 for a table that has only 6 rows remaining needs to be reduced to 6... BUG FIXES - Tables with a lot of COLSPANs could cause a divideby-zero error or bad pages (NAN instead of a number.) - Table cells with a single render element would not be vertically aligned. - The --quiet option would enable progress messages on the command-line. - Table cell widths could be computed incorrectly, causing unnecessary wrapping. - The --path option disabled checking for the file with the original filename. CHANGES IN HTMLDOC v1.8.13

NEW FEATURES - Added support for secure (https) URLs via the OpenSSL library. - Added support for Acrobat 5.0 (PDF 1.4). - Added support for transparency in PostScript and PDF 1.1 and 1.2 output. - Added a --no-jpeg option (same as --jpeg=0) - Added support for the CSS2 page-break-before and page-break-after properties. - Added a PHP example. CHANGES - External file references to non-PDF files now use the "Launch" action so they can be opened/executed/saved as allowed by the OS and PDF viewer. - Changed the indexed/JPEG'd transition point to 256 colors when using Flate compression. This makes PDF files much smaller in general. - Changed the in-line image size limit to 64k. - Now allow "<" followed by whitespace, "=", or "<". This violates the HTML specification, but we're sick of people complaining about it. - Preferences are now stored in a user-specific file under Windows, just like UNIX. This provides user-specific preferences and allows preferences to be kept when upgrading to new versions of HTMLDOC. - The book loading code now allows for blank lines, even though these are not a part of the format. (added to support some scripted apps that include extra newlines...) - Changed the leading space handling of blocks to more closely match the standard browser behavior. BUG FIXES - The table formatting code adding the border width to the cell width, while Netscape and MSIE don't. This caused some interesting formatting glitches... - The table formatting code didn't account for the preferred width of colspan'd cells. - The table formatting code tried to enforce the minimum cell width when squeezing a table to fit on the page; this caused the table to still exceed the width of the page. - The PDF catalog object could contain a reference to a /Names object of "0 0 R", which is invalid. This would happen when the "--no-links" option was used. - Several HTML elements were incorrectly written with closing tags. - When piping PDF output, the temporary file that is created needed to be open for reading and writing, but HTMLDOC only opened the file for writing. - Image links did not work. - The JPEG image loading code did not correctly handle grayscale JPEG images. - JPEG images were not encrypted when writing a document with encryption enabled.

- The user password was not properly encrypted. - The colormap of indexed images were not encrypted when writing a document with encryption enabled. - The temporary file creation and cleanup functions did not use the same template under Windows, causing multiple conversions to fail when temporary files were used. - Paragraphs could end up with one extra text fragment, causing the line to be too long. - The command-line program would clear the error count after reading all the files/URLs on the command-line, but before generating the document. If there were problems reading the files/URLs, HTMLDOC would return a 0 exit status instead of 1. - Image objects that were both JPEG and Flate compressed would not display (filters specified in the wrong order.) - Images with more than 256 colors would cause a segfault on some systems. - Background images would generate the error message "XObject 'Innn' is unknown". CHANGES IN HTMLDOC v1.8.12 NEW FEATURES - Added new "--batch" option to convert HTMLDOC book files from the command-line. - Added support for the "-display" option on systems that use X11. - Now use image objects in PDF output for images when the image width * height * depth > 32k. - Now use JPEG compression when the number of colors would be > 32 colors or 16 gray shades. - True transparency support for GIF files in PDF 1.3 output! - The GUI now automatically changes the extension of the output filename as needed. - The GUI now collects all error messages and shows them once after the document is generated. - Added support for HSPACE and VSPACE attributes for images with ALIGN="LEFT" or ALIGN="RIGHT". - Added new Java interface to HTMLDOC. CHANGES - Consolidated temporary file management into new file_temp() function. The new function also makes use of the Windows "short lived" open option which may improve performance with small temporary files. - Updated book file format and added an appendix describing the format. - Now default to PDF 1.3 (Acrobat 4.0) output format. - Now output length of PDF streams with the stream object; this offers a modest reduction in file size. - The HTTP file cache now keeps track of previous URLs that were downloaded. - The HTTP code now supports redirections (status codes 301 to 303) to alternate URLs.

- Limit the height check for table rows to 1/8th of the page length; this seems to provide fairly consistent wrapping of tables without leaving huge expanses of blank space at the bottom of pages. - The HTML output now also includes a font-family style for PRE text; otherwise the body font would override the PRE font with some browsers. - The snprintf/vsnprintf emulation functions were not included in the HTMLDOC makefile. - RGB hex colors are now recognized with or without the leading #. This breaks HTML standards compliance but should reduce the number of problem reports from buggy HTML. - The stylesheet generated with the HTML output no longer contains absolute font sizes, just the typefaces and a relative size for SUB/SUP. - The title image is no longer scaled to 100% in the HTML output. BUG FIXES - The web page output was not divided into chapters for each input file. - The "make install" target did a clean. - The configure script would remove the image libraries if you did not have FLTK installed. - The fix_filename() function didn't handle relative URLs for images (e.g. SRC="../images/filename.gif") - Comments in the source document were being closed by a </!--> tag in the HTML output. - The alignment attribute in TR elements was not inherited by the TD and TH elements. - The HTML parser added whitespace before the title of a document (missing check for TITLE element.) - The table formatter did not reset the column width when a width was not specified explicitly. This caused the columns to be formatted with equal widths... - Paragraphs that didn't use the P element would use the alignment attribute of the first fragment instead of the parent. CHANGES IN HTMLDOC v1.8.11 NEW FEATURES - Added koi8-r character set. - Added new "TrueType" font option for PDF output. This (hopefully) should improve support for non-latin languages. - Added support for "justify" alignment; this is currently implemented by adding additional space between characters (no automatic hyphenation...) CHANGES - The "make install" target does a "make all" to ensure that the software is built before installing it. This should help users that don't read the documentation build the software.

- Incorporated several OS/2 compile fixes from Alexander Mai. - Tables that exceed the printable width of the page are now squeezed to fit. BUG FIXES - The temporary file created for PDF output to stdout was not unique. - The temporary file created for PDF output to stdout did not use the GetTempPath() function under Windows. - The temporary file cleanup code did not use the GetTempPath() function under Windows. - The prefs_load() function did not check the HTMLDOC_DATA environment variable until after loading the preferences file. This could cause any saved settings to generate error messages about missing files (these were reloaded when the document data was read, however.) - The first border for a table cell that spanned multiple pages did not account for the cellpadding or border width. - Leading whitespace was not eliminated in all cases. CHANGES IN HTMLDOC v1.8.10 NEW FEATURES - New "--quiet" option to suppress all messages sent to stderr. - New chapter n/N page number format (:) - New "--links" and "--no-links" options for PDF output. - Added "&euro;" character name support. CHANGES - Documentation updates. - If a heading already contains a link, the name assigned for the heading is stored in the existing link to avoid nested links. - The table parsing code now also traverses THEAD and TFOOT sections and handles multiple TBODY, THEAD, and TFOOT sections. The THEAD and TFOOT rows are treated as ordinary rows. - Image alignment should now match W3C recommendations. - Newlines inside quoted values (e.g. SRC="filename\r\n") are now ignored. - The STYLE data generated in HTML output now includes the TYPE="text/css" attribute. BUG FIXES - The path support was still broken. - Embedded files and images did not use path or HTTP support. - Table cell borders could be drawn on the wrong page if not all cells in a row span more than one page. - Large images could end up covering the footer at the bottom of the page (wasn't accounting for the line

spacing.) - Fixed some memory leaks that would affect Windows 95/98/Me users. CHANGES IN HTMLDOC v1.8.9 NEW FEATURES - Added a "--no-numbered" option to turn heading numbers off. - Added support for "keywords" META data. - Added support for BMP images. - Added support for ROWSPAN attribute in tables. - Added support for HTTP file references. - Added new sample CGI program that can produce a PDF file for any page on a server. - Added new n/N, date, and time formats for the header and footer. CHANGES - Configuration script changes. - Now ignore file count in book files; instead, we now look at the first character of the third and subsequent lines - a dash (-) indicates the start of the options. (use a backslash to quote filenames starting with a dash) - Multiple-line paragraphs that have only 1 line on the bottom of the current page are now started on the following page. - DSC comments in PostScript output were not 100% conformant with the DSC spec. - Long table-of-contents entries are now wrapped (original patch supplied by Richard Pennington) - New color icon under UNIX when Xpm library available. BUG FIXES - Didn't allow &#nnn; character escapes inside preformatted text. - Empty TBODY groups would cause parse_table() to crash. - Comments were incorrectly terminated by ">" instead of "-->". - The command-line and GUI interfaces looked for "outlines" instead of "outline" for the page mode. - The HTML output code didn't output closing tags for empty elements. - The GUI interface started with the compression slider enabled, even for HTML output. - The beginnings of some lines could start with whitespace. - Wasn't aligning images and text on lines based on the line height. - The compression slider was enabled in the GUI even though HTML output was selected. - The Perl example code was incorrect. - Fixed the check for whether or not pages were generated.

- htmlSetCharSet() wasn't reloading the character set data if the data directory changed. - The GUI did not reset the default background color. - The 'C' page number style (chapter page numbers) started at 3 instead of 1. - The chapter links were off by 1 or 2 pages when no title page was included. CHANGES IN HTMLDOC v1.8.8 NEW FEATURES - Added support for PDF security/encryption! - Now support TABLE height attribute. - Now generate an error message if no pages are generated (with a suggestion to use the webpage option.) - New "paths" option to specify additional directories to search for files. This is useful when the source files use absolute server paths. CHANGES - Added missing casts in htmllib.cxx that were causing a compile warning with some compilers. - No longer draw borders around empty cells in tables.. - Now disable the TOC tab when using webpage mode. - Now scale title image to 100% in HTML output. - Now handle comments with missing whitespace after the "<!--". BUG FIXES - Nested tables didn't take into account the table border width, spacing, or padding values. - HTMLDOC crashed under Solaris when reading HTML files from the standard input. - <ELEM>text</ELEM> <MELE>text</MELE> was rendered without an intervening space. CHANGES IN HTMLDOC v1.8.7 NEW FEATURES - The configure script now uses the local PNG, ZLIB, and/or JPEG libraries when they are new enough. - The configure script now uses the -fno-rtti, -fno-exceptions, and -fpermissive options as needed with GCC (smaller, faster executables, works around X header bugs in Solaris.) - Added a --toctitle option to set the table-of-contents title from the command-line (was only available in the GUI in previous releases...) - New "<!-- NEED amount -->" comment to force a page feed if there is not sufficient room on the page for the following text. - Page comments are now supported in tables. - Table rows are now allocated dynamically, MAX_ROWS at

a time. CHANGES - Increased default MAX_PAGES to 10000 (was 5000.) - File links in book files now point to the top of the next page. - <TABLE ALIGN=xyz> now aligns the table (previously it just set the default alignment of cells.) - Transparent GIFs now use the body color instead of white for the transparent color. - Updated to LIBPNG 1.0.6 in source distribution. - Updated the default cellpadding to be 1 pixel to match Netscape output. - Updated line and block spacing to match Netscape. - DL/DT/DD output now matches browsers (was indented from browser output.) - Now only output link (A) style if it is set to "none". Otherwise Netscape would underline all targets as well as links. - Increased the MAX_COLUMNS constant to 200, and dropped MAX_ROWS to 200. Note that the new table code now allocates rows in increments of MAX_ROWS rows, so the actual maximum number of rows depends on available memory. BUG FIXES - Now ignore illegal HTML in tables. - The VALIGN code didn't handle empty cells properly. - Wasn't offsetting the start of each row by the cell padding. - The JPEG image loading code didn't work for some JPEG images, particularly those from digital cameras (JPEG but not JFIF format.) - The strikethrough line was not being drawn in the correct position. - Wasn't setting the height of BR elements, so <BR><BR> didn't insert a blank line. - The table of contents would show the wrong page numbers if no title page was generated. - Cell widths did not subtract any border, padding, or spacing from the "preferred" width, causing formatting differences between web browsers and HTMLDOC. - The PNG loading code did not handle interlacing or transparency. - The HTML parsing code did not prevent elements in embedded files from completing elements in the parent file. - The table CELLSPACING amount was being applied twice in the table sizing calculations. CHANGES IN HTMLDOC v1.8.6 NEW FEATURES - New linkcolor and linkstyle options. CHANGES

- Minor source changes for OS/2 compilation. - SUP and SUB now raise/lower text more to be consistent with browser look-n-feel. - Non-breaking space by itself was being output. Now check for that and ignore strings that consist entirely of whitespace. - New progress bar. BUG FIXES - Didn't add whitespace after a table caption. - Nested tables caused formatting problems (flatten_tree() didn't insert breaks for new rows) - A cell whose minimum width exceeded the available width for the table would cause the table to go off the page. - Cells that spanned more than two pages were drawn with boxes around them rather than just the sides. - The stylesheet info in the HTML output specified the H1 size for all headings. - The title page was incorrectly formatted when an image was specified - the text start position was computed using the pixel height of the title image and not the formatted height. - 1 color images didn't come out right; the "fix" to work around an Acrobat Reader bug was being done too soon, so the color lookups were wrong. - HTML file links now work properly. - Now limit all HTML input to the maximum size of input buffers to avoid potential buffer overflow problems in CGIs. - If a row had a predefined height, HTMLDOC wasn't making sure that the row would fit on the current page. - THEAD, TFOOT, and TBODY caused problems when formatting tables. Note: THEAD and TFOOT are *still* not supported, however the code now properly ignores them and parses the rows in the TBODY group. - The VALIGN code introduced in the 1.8.5 release didn't check for NULL pointers in all cases. CHANGES IN HTMLDOC v1.8.5 NEW FEATURES - New "--titlefile" option to include an HTML file for the title page(s). - New 'C' header/footer option to show current page number within chapter or HTML file. - Allow adding of .book files to import all HTML files in the book. - New "HALF PAGE" page comment to feed 1/2 page. - Added VALIGN and HEIGHT support in tables. CHANGES - Now optimize link objects in PDF files (provides a 40k reduction in file size for the HTMLDOC manual alone) - Table rows that cross page boundaries are now rendered more like Netscape and MSIE. - Now support HTMLDOC_DATA and HTMLDOC_HELP environment variables under UNIX (for alternate install directory) - Now show error messages when HTMLDOC can't open the

AFM, character set, or PostScript glyph files. - The logo image is now scaled to its "natural" size (as it would appear in a web browser) - Now recognize VALIGN="MIDDLE" or VALIGN="CENTER". BUG FIXES - Generation of PDF files to the standard output (i.e. to the web server + browser) didn't work on some versions of UNIX. HTMLDOC now writes the PDF output to a temporary file and then copies it to the standard output as needed. - PDF links were missing the first 5 characters in the filename; the code was trying to skip over the "file:" prefix, but that prefix was already skipped elsewhere. - Nested descriptive lists (DL) did not get rendered properly. - Tables had extra whitespace before and after them. - Multiple aligned images confused parse_paragraph(); the images would overlap instead of stack on the sides. CHANGES IN HTMLDOC v1.8.4 CHANGES - More configure script changes for FLTK DSOs. - FileIcon.cxx was still using NULL for outline (an integer), which caused some ANSI C++ compilers to complain. BUG FIXES - The Fonts and Colors tab groups did not extend to the full width of the tab area, which prevented the Browse button from working when clicked on the right side. - The help dialog window did not scroll all the way to the bottom of the text. - The chapter ("c") header/footer string did not work. - The heading ("h") header/footer string did not always match the first heading on a page. - The header and footer fonts were not used when computing the widths of the header and footer strings. - The Windows distribution did not create the right shortcut for the Users Manual in the Start menu. - The command-line code did not accept "--grayscale", only "--gray" - Multi-file HTML output did not use the right link for the table-of-contents file if no title page was being generated. - Extra whitespace before and after tables has been eliminated. CHANGES IN HTMLDOC v1.8.3 NEW FEATURES

- New "--browserwidth" option to control scaling of images and tables that use fixed pixel widths. CHANGES - The configure script now looks for the OpenGL library (required if you use a shared FLTK library with OpenGL support.) - Increased the max number of chapters to 1000. BUG FIXES - Page break comments didn't force a paragraph break. - --no-toc prevented chapters from being output in PS and PDF files. - Filenames didn't always get updated properly when doing a "save as"... - Fixed some more leading/trailing whitespace problems. - Wasn't freeing page headings after the document was generated. - Wasn't range checking the current chapter number; now limits the number of chapters to MAX_CHAPTERS and issues an error message whenever the limit is exceeded. CHANGES IN HTMLDOC v1.8.2 NEW FEATURES - New "setup" program for UNIX software installation. CHANGES - Documentation updated for new UNIX "setup" program and "..." usage for headers and footers. - Changed margins to floating point (instead of integer) to improve table column accuracy. BUG FIXES - HTMLDOC could crash under Microsoft Windows with some types of HTML files. This was caused by a stack overflow, usually when processing nested tables. - Multiple HTML files weren't being converted properly in web page mode - only the last file would be generated for PostScript output, and no file for PDF output. - Wasn't preserving the whitespace between "one" and "two" in the HTML code "one<I> two</I> three". - Paragraph spacing was inconsistent. - <TABLE WIDTH="xx"> wasn't formatted properly. - The command-line code wasn't opening HTML files in binary mode. This caused problems under Microsoft Windows. CHANGES IN HTMLDOC v1.8.1 CHANGES - The configure script didn't update the ARFLAGS variable for *BSD operating systems (no "s" option to

build the symbol table...) - Changed the installation commands to only create the installation directory if it does not exist. This prevents installation errors on some platforms the second time around. - Now use the Microsoft definitions for characters 128 through 159 that are otherwise unused by the ISO-8859-x character sets. - Now set optimization settings when we know the compiler. - Now always quote attribute values in HTML output to make HTML lint programs happy. BUG FIXES - Wasn't using TOC title string in PDF document outline. - Preformatted text in tables didn't force the column width. - Cells using COLSPAN > 1 didn't contribute to the width of columns. - The table code didn't enforce the per-column minimums under certain circumstances, causing "scrambled" columns. - The configure script and makefiles didn't work when FLTK was not available. They now only build the "gui" library when it is available. - The Windows distribution was installing files under PROGRAMDIR instead of TARGETDIR. This prevented users from customizing the installation directory. - The configure script overrode the LDFLAGS environment variable, preventing FLTK from being located in a nondefault directory. CHANGES IN HTMLDOC v1.8 NEW FEATURES - Now support PDF 1.1 (Acrobat 2.x) and PDF 1.3 (Acrobat 4.0). - Now support PDF page modes, layouts, and effects, and the first page that is displayed in Acrobat Reader. - Now support PostScript Level 3 output with Flate image compression. - Now support PostScript commands for page size and duplexing. - Now add filenames as needed to HTML links. - Added optimizations to output code to further reduce PDF and PostScript file size. - Now support alternate 8-bit character sets. Currently we supply data files for the ISO-8859-N character sets. - Added chapter headings to the available header/footer formats. - The GUI file chooser is significantly improved and supports selection of multiple HTML files. - The GUI now provides on-line help. - Many other GUI improvements. - Added support for DIR and MENU block elements. - The header and footer text can now be made boldface, italic, etc. - Font settings are now exported to HTML files in a style sheet. - Now support page breaks using HTML comments.

The image dimensions are now exported to HTML files. Added landscape printing option. Added CAPTION support for tables. Filename links now work for HTML files included in a document. - Now support BGCOLOR in tables. CHANGES - Lots of documentation changes. - Much better table formatting. - Changed HTML output to use less invasive navigation bars at the top and bottom of each file. This also means that the "--barcolor" option is no longer supported! - Updated to use existing filenames in HTML (directory) output. - Now recognize any local PDF file as a local file link (i.e. you just need "HREF=filename.pdf" and not "HREF=file:filename.pdf") - <TT>, <CODE>, and <SAMP> no longer reduce the font size. - Now put whitespace after image data in PDF files. This change was needed to work around a bug in Acrobat Reader 4.0. - Now generate a complete encoding vector for fonts in PDF files. This change was needed to work around a bug in all versions of Acrobat Exchange that did not recognize the WinANSI encoding defined in the PDF specifications. - Now filter out the BREAK attribute from HR elements. - Now only load images once. BUG FIXES Wasn't escaping &,<, or > in HTML output Wasn't preserving &nbsp; Links in multi-file HTML output were off-by-one. BLOCKQUOTE needed to be like CENTER and DIV. Needed to use existing link name if present for headings to avoid nested link name bug in Netscape and MSIE. Extremely long link names could cause TOC generation to fail and HTMLDOC to crash. PDF output was not compatible with Ghostscript/Ghostview because Ghostscript does not support inherited page resources or the "Fl" abbreviation for the "FlateDecode" compression filter. PostScript DSC comments didn't have unique page numbers. This caused Ghostview (among others) to get confused. Some functions didn't handle empty text fragments. Images couldn't be scaled both horizontally and vertically. <LI> didn't support the VALUE attribute (but <OL> did...) Fixed whitespace problems before and after some markups that was caused by intervening links. The indexed image output code could generate an image with only 1 color index used, which upset Acrobat Reader. Fixed a bug in table-of-contents handling - HTMLDOC would crash on some systems if you converted a web page on the command-line. Wasn't setting the font size and spacing soon enough when generating files on the command-line. Didn't hide EMBED elements when generating indexed HTML files. Didn't always set the current drawing position before drawing a box or line. Base85 encoding of image data was broken for PostScript output. JPEG compression was broken for PostScript output.

- Didn't set binary mode for the standard output under Windows and OS/2 needed. CHANGES IN HTMLDOC v1.7 Added option for overriding the background color or image. Added default font typeface and size options. Added progress indicator for page formatting. The HTMLDOC window is now resizeable. The <TABLE> and <CENTER> markups didn't start a new block. strcasecmp and friends are not available on all platforms. Added support for MacOS (command-line only). The width of table cells could be off by 1 point causing unnecessary text wrapping. The GUI's default center footer wasn't "blank". Images could be "lost" if they reside in the current directory or use an absolute path. Documents without titles or headings could crash HTMLDOC. The image loading code could crash due to a MSVC++ runtime library bug. Spacing before <HR>'s wasn't consistent. Buffer overflow problems causing crashes. Didn't accept whitespace in variables, e.g. "<TAG NAME = VALUE>" Links didn't always get propagated. The Flate compressor data was not getting freed, so HTMLDOC could use a lot of memory when compression was enabled.

CHANGES IN HTMLDOC v1.6 - Now support JPEG compression of images. - Now have selectable Flate (ZIP) compression level. - Now only adjust top and bottom margins if headers and footers are used. - Better HTML output support (now remember files for links in multi-file output). - Increased maximum page count to 5000. - Needed to show headers on all pages in web page mode. - Now recognize both "in" and "inch" for measurements. - <BR> was not handled properly. - Selecting "web page" in the GUI clears the title toggle. - TABLE row spacing was not right... - <TD COLSPAN=n> now draws multi-column borders. - Column widths were computed wrong when COLSPAN was used. - Nested lists were not handled right. - Internal links didn't work for PDF output. - Block spacing should now be more consistent. - Image scaling was off - now only use page width so that images are not warped. - The footer was always one line too low. - Couldn't double-click on input filename to edit. CHANGES IN HTMLDOC v1.5 Added customization of headers and footers. Added new "--title" image option. Can now put logo image in header or footer. <MARKUP ID="name"> now works for link destinations.

- The table of contents now appears as part of the document outline in PDF output. - Links to local PDF files are now treated as file links in PDF output instead of web links. - You can now turn the title page on/off as desired. - PostScript and PDF output to stdout now works. - Nested tables now format properly. - <HR> now provides horizontal rule; to get a page break use "<HR BREAK>". - Fixed <TABLE BORDER=0> bug. - Fixed GIF loader bug (caused problems on Alpha machines) - No longer get extra line after list items. - <FONT> markup nesting now works. - "&" by itself would cause loss of 15 characters. - The current directory was not tracked properly under Windows. - The right, top, and bottom margins were not being saved properly. - The htmlReadFile() function could consume too much stack space, causing a program failure. - PostScript and PDF files were corrupt when generating a web page with a title page. CHANGES IN HTMLDOC v1.4 - Now use autoconf "configure" script to build UNIX makefile. - Now handle relative filenames a lot better when loading images and files. - Added "--webpage" option to support printing of plain HTML files (i.e. not documents with chapters) - Added support for document backgrounds in PostScript and PDF output - Added "--no-toc" and "--no-title" options to disable the table-of-contents and title pages, respectively - PDF files now store all named links for use from a web page (HREF="filename.pdf#name") - Converted to C++ - Now using FLTK for the GUI under UNIX and Windows (yeah, one set of code!) - Merged GUI and command-line versions - Greatly enhanced GUI now supports nearly all command-line options. - Miscellaneous fixes to HTML parsing code - PDF links should now go to the right page all the time - Fixed DSC comments in PostScript output to conform to the standard - Fixed dumb bug in Windows version - didn't handle HTML files with only a LF at the end of each line (this is a BUG in the MSVC++ runtime libraries!) - <PRE> inside a list didn't work - parse_table() and friends didn't check for a NULL parent pointer. - Paragraph text that wasn't enclosed by P markups was located on the wrong page when followed by a H1 markup. CHANGES IN HTMLDOC v1.3.1 - Fixed font encoding vector in PostScript output (minus instead of hyphen for '-' character).

CHANGES IN HTMLDOC v1.3 - New GUI for managing documents (Windows + X11/Motif) - Better table printing with support for user-specified column widths and better automatic-sizing - PNG loading now works when grayscale output is requested - No image optimization was performed in PDF or Level 2 PostScript files. HTMLDOC now converts images to indexed (1,2,4,8 bits) if there is an advantage (fewer bits per pixel) and no loss of color would occur - The filenames in links were getting lost when writing indexed HTML to a directory - The logo image filename wasn't being localized when writing indexed HTML to a directory - Fonts, images, and links weren't supported inside a PRE tag - Added support for the <!DOCTYPE> markup - No longer assume that chars are unsigned by default - Invalid or missing links no longer generate bad PDF files - External links (http:, ftp:, etc) now work - Escaped characters are now decoded correctly in the table of contents in PDF files - Image scaling is now more intelligent CHANGES IN HTMLDOC v1.2 - Now support "internal" links in a document (PDF & HTML). - Added "no compression" option for PDF files; this is needed for older PDF readers like Acroread 2.x. - Much better parsing of HTML; should now work very well with the HTML output by Netscape Composer. - Wasn't opening image files in "binary" mode (Windows). - The htmlReadString() and htmlWriteString() functions were removed because of portability problems to HP-UX and Windows, among others. CHANGES IN HTMLDOC v1.1 - Ordered (numbered) lists are now supported, as are the TYPE=, START=, and VALUE= option variables. - Now support coverpages for PS and PDF output with optional logo image. - Running headings (at the bottom of PS/PDF pages) are now tracked correctly. - Fixed parsing of lists so lists generated by Netscape Composer work right... - Fixed HTML links when generating a single HTML file. - The --numbered option didn't number all headings (only those in the table-of-contents). CHANGES IN HTMLDOC v1.0 - Initial version.